With the rapid development of network information technology and internet technology, more and more people choose online learning as an important channel to acquire knowledge.However, while online education is flourishing, users are facing problems such as inaccurate recommendations and imbalanced paths caused by information overload and data silos.Therefore, this article proposes the Federated Union Genetic Recommendation (FUGR) model, which aggregates cross school data using longitudinal federated learning and evolves multi-objective weights using association rule-based genetic algorithms to achieve accurate recommendation of learning resources required by users.This article conducted a series of experiments on the proposed model, and the results showed that FUGR improved the baseline AUC by 7.5%, HR by 6.4%, path consistency by 22%, scarce course coverage by 15%, while maintaining a privacy budget of < 1.
Yumei Zhang (Thu,) studied this question.